Towards Efficient Graph Traversal using a Multi-GPU Cluster

Abstract

Graph processing has always been a challenge, as there are inherent complexities in it. These include scalability to larger data sets and clusters, dependencies between vertices in the graph, irregular memory accesses during processing and traversals, minimal locality of reference, etc. In literature, there are several implementations for parallel graph processing on single GPU systems but only few for single and multi-node multi-GPU systems. In this paper, the prospects of improvement in large graph traversals by utilizing multi-GPU cluster for Breadth First Search algorithm has been studied. In this regard, a DiGPU, a CUDA-based implementation for graph traversal in shared memory multi-GPU and distributed memory multi-GPU systems has been proposed. In this work, an open source software module has also been developed and verified through set of experiments. Further, evaluations have been demonstrated on local cluster as well as on CDER cluster. Finally, experimental analysis has been performed on several graph data sets using different system configurations to study the impact of load distribution with respect to GPU specification on performance of our implementation.

Authors and Affiliations

Hina Hameed, Nouman M Durrani, Sehrish Hina, Jawwad A. Shamsi

Keywords

Related Articles

Smart Grid Network Transmission Line RLC Modelling Using Random Power Line Synthesis Scheme

This work proposes Random Power line Synthesis (RPLS) as a quicker computational approach to solving RLC parameters of a modern smart grid transmission network. Since modern grid systems provide a holistic perspective of...

Improving Throughput and Delay by Signaling Modification in Integrated 802.11 and 3G Heterogeneous Wireless Network

Current trends show that UMTS network and WLAN will co-exist and work together to support more users with higher data rate services over a wider area. However, this integration invokes many challenges such as mobility ma...

How to Improve the IoT Security Implementing IDS/IPS Tool using Raspberry Pi 3B+

This work shows a methodology of implementation and testing of the system is proposed and tested with a prototype; it is constructed with sensors and actuators that allow monitoring the behavior of the system in an envir...

Quality of Service Provisioning in Biosensor Networks

Biosensor networks are wireless networks consisting of tiny biological sensors (biosensors, for short) that can be implanted inside the body of human and animal subjects. Biosensors can measure various biological process...

LASyM: A Learning Analytics System for MOOCs

Nowadays, the Web has revolutionized our vision as to how deliver courses in a radically transformed and enhanced way. Boosted by Cloud computing, the use of the Web in education has revealed new challenges and looks for...

Download PDF file
  • EP ID EP259649
  • DOI 10.14569/IJACSA.2017.080644
  • Views 101
  • Downloads 0

How To Cite

Hina Hameed, Nouman M Durrani, Sehrish Hina, Jawwad A. Shamsi (2017). Towards Efficient Graph Traversal using a Multi-GPU Cluster. International Journal of Advanced Computer Science & Applications, 8(6), 338-346. https://europub.co.uk/articles/-A-259649